Motion Artifact Simulation for Medical Imaging Training Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for reducing motion artifacts in scanning images of moving objects, such as the heart, require a large number of images with motion artifacts for training models, which is challenging to obtain.
Innovation Solution
A system and method for simulating motion artifacts by determining motion vector fields and reconstruction images over sub-periods of a time period, using a target image and an artifact simulation model to generate a motion artifact simulation image, which can be used to train a motion artifact removal model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a motion artifact removal model is trained using real scanning images containing motion artifacts, then the model can correct motion artifacts in scanning images, but a large number of real images containing motion artifacts are required for training which are difficult to obtain
Solution Approach 1:
The patent creates synthetic copies of real scanning images by generating motion artifact simulation images through a simulation model. This allows the training data to be replicated without requiring additional real patient scans, solving the scarcity problem while maintaining training quality through realistic artifact simulation
Solution Approach 2:
The patent transforms parameters of the simulation model (such as motion vector fields, artifact types, and intensity levels) to generate diverse training images. By varying these parameters systematically, the system can produce a large quantity of training images with different motion artifact characteristics from a single real scan
2Productivity
If more real scanning images with motion artifacts are collected to train the model, then the model performance improves, but the time and resources required to collect and process these images increases
Solution Approach 1:
The patent pre-processes real scanning images by extracting motion vector fields and other motion information before generating synthetic training images. This preliminary extraction of motion characteristics allows rapid generation of multiple training images from each real scan, significantly reducing the time required compared to collecting new real images
Solution Approach 2:
By creating synthetic copies of real images through the simulation model, the system can rapidly generate large quantities of training data without the time-consuming process of collecting actual patient scans. The copying process uses extracted motion information to efficiently reproduce artifacts in synthetic images
Data Source
AI summary
Systems and methods for motion artifact simulation are provided. The systems may obtain a target image including a target object. The systems may determine a plurality of sub-periods of a time period corresponding to the target image. The systems may determine a plurality of motion vector fields of the target object in the plurality of sub-periods. Each motion vector field of the plurality of motion vector fields may correspond to one of the plurality of sub-periods. The systems may determine a plurality of reconstruction images of the target object corresponding to the plurality of sub-periods based on projection data of the target image. Each reconstruction image of the plurality of reconstruction images may correspond to one of the plurality of sub-periods. The systems may generate a motion artifact simulation image of the target object based on the plurality of motion vector fields and the plurality of reconstruction images.